The wealth of information provided by whole-exome and whole-genome sequencing studies can make it difficult to identify variants that cause disease. Family sequencing, in particular the sequencing of trios and quartets with affected children and unaffected parents, allows one to exclude most of the rare variants present in the genome and focus on those few that are de novo or rare and homozygous only in the offspring. There is growing recognition that this is a productive approach in the study of complex diseases. Here, we propose to perform whole exome sequencing in families to identify genetic variants predisposing to obsessive-compulsive disorder (OCD). OCD is among the 10 most disabling medical conditions worldwide and involves persistent, intrusive, senseless thoughts and impulses (obsessions) and repetitive, intentional behaviors (compulsions). The lifetime prevalence of OCD is estimated at 1-3%. Although medication and behavioral therapy, available since the late 1980s, are useful, they control symptoms with only limited success, and the course of the condition remains chronic in most cases; cure is rare. It is therefore critical to understand the pathophysiology of OCD so we can ultimately develop effective treatments. We propose to sequence the exomes of 375 simplex OCD trios and 100 OCD quartets comprising affected sibs and unaffected parents. In the trios, we will identify de novo variants that are annotated as functional. Even if de novo variants are found to explain only a small proportion of OCD cases, their discovery will provide a substantial advance in understanding the pathophysiology of the disorder. Furthermore, we will analyze the distribution of de novo variants found in the trios, comparing this distribution to that expected in the general population and that found in trios from other psychiatric diseases. In the quartets as well as the trios, we will identify causal variants that are homozygous or compound heterozygous in the offspring but heterozygous in the parents. We will identify candidate variants falling into these categories and will prioritize them based upon functional annotation, rarity and overlap of implicated genes between unrelated families. We will then confirm these variants by genotyping them in unaffected family members. Finally, we will identify genotype-phenotype correlations by looking for characteristics that are shared between unrelated cases with causal variants in the same gene and also by looking for genetic similarities within clearly defined clinical subgroups of patients, such as those with tic disorders.